Canvas Mcp
by vishalsachdevcommunityv1.2.0
96
32
MCP server for Canvas LMS that lets AI assistants read and manage course, assignment, discussion, page, module, grading, and accessibility-related data through natural language. It supports student, educator, learning designer, and developer workflows with role-based tool filtering and bulk operations.
From the registry: Canvas LMS integration for students and educators with FERPA-compliant analytics and workflows
Install & configure
Prerequisites
- requiredPython 3.10+Python 3.10 or higher is required
- requiredInstall via pipInstall canvas-mcp from PyPI
https://pypi.org/project/canvas-mcp/$ pip install canvas-mcp - requiredGet a Canvas API tokenCanvas → Account → Settings → New Access Token
Install in your client
Paste into the AI you're chatting with — it will figure out the rest
Please install the `canvas-mcp` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Python 3.10+** — Python 3.10 or higher is required
- **Install via pip** — Install canvas-mcp from PyPI Run: `pip install canvas-mcp` (https://pypi.org/project/canvas-mcp/)
- **Get a Canvas API token** — Canvas → Account → Settings → New Access Token
Canonical MCP server config (stdio transport):
- command: `canvas-mcp-server`
- args: []
- required environment variables:
- `CANVAS_API_TOKEN`: Your Canvas LMS API access token (example: `<your-canvas-api-token>`)
- `CANVAS_API_URL`: Your institution's Canvas API URL (e.g., https://canvas.illinois.edu/api/v1) (example: `<your-canvas-url>/api/v1`)
- optional environment variables:
- `CANVAS_ROLE`: Filter tools by role: student, educator, or admin (example: `student`)
- `ENABLE_DATA_ANONYMIZATION`: Anonymize student PII for FERPA compliance (example: `true`)
- `LOG_REDACT_PII`: Redact PII from logs (example: `true`)
- `LOG_ACCESS_EVENTS`: Enable structured audit logging (example: `false`)
- `ENABLE_TS_SANDBOX`: Enable TypeScript execution sandbox (example: `true`)
Note: 87 tools for Canvas LMS. A hosted version is also available at https://mcp.illinihunt.org/mcp (HTTP) requiring X-Canvas-Token and X-Canvas-URL headers.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.Note: 87 tools for Canvas LMS. A hosted version is also available at https://mcp.illinihunt.org/mcp (HTTP) requiring X-Canvas-Token and X-Canvas-URL headers.
Tools (39)
get_my_upcoming_assignments
Due dates for the next N days
get_my_todo_items
Canvas TODO list
get_my_submission_status
Check submitted vs missing assignments
get_my_course_grades
Get current grades
get_my_peer_reviews_todo
List pending peer reviews
list_assignments
List all assignments in a course
create_assignment
Create a new assignment
update_assignment
Update an existing assignment
list_submissions
List student submissions
bulk_grade_submissions
Grade multiple submissions at once
get_assignment_analytics
Show assignment performance statistics
send_conversation
Message students
create_announcement
Post announcements
create_module
Create a course module
update_module
Environment variables
CANVAS_ROLErequiredSets the visible tool set to student, educator, or adminENABLE_DATA_ANONYMIZATIONrequiredEnables local data anonymization for student dataPrerequisites
- • Python 3.10+
- • python3
Packages
canvas-mcp
pypiv1.2.0transport: stdio
Details
- Language
- Python
- License
- MIT
- Released
- Apr 10, 2026
- Last commit
- Apr 12, 2026
- Registry name
- io.github.vishalsachdev/canvas-mcp